different mechanisms, as those shown in Figs.4 and 5. On the basis
of the latter one, in particular, it looks very promising the possible
synergic combination of a multi-agent environment with other
general-purpose programmable simulation algorithms (like genetic
algorithms) to underpin the basic mechanisms at the root of highly
complex functions in the human Central Nervous System.
In any case, a common mechanism for migraine and epilepsy
based upon the synchronization of specific neuronal regions was
confirmed and satisfactorily simulated also by a multi-agents pro-
gramming environment, which paves the way for further, more
ambitious extensions [19].3 Functional Connections and Correlations in Brain Networks
Using a MAS-based simulation engine to study the functional
properties of a neuronal network is simplified by the straightfor-
ward identification of MAS agents and links with, respectively,
network nodes and ties. In a MAS representation of neuronal net-
works, in fact, an agent can be assimilated to a single neuron, to a
set of functionally correlated neurons, or even to a given brain
region. Moreover, functional relationships among sub-networks
are not necessarily associated with well-defined anatomical
connections.Fig. 4Synchronized activity in brain neurons. Thetop left,middle,andright panelsshow the active neurons at
the maximum, minimum, and intermediate synchronization levels, respectively. Thebottom panelsshow the
corresponding trend of metabolic energy. Notice, close to the central sulcus in the right hemisphere, the
“epileptic focus” triggering the synchronization wave
314 Alfredo Colosimo